基于改进PID神经元网络的多变量系统控制算法

PID神经元网络具有动态特性，在系统控制应用中相比于传统的PID控制方法可取得更优的效果，但其学习算法为梯度学习算法，初始权值随机取得，为了提高其控制量逼近控制目标的速度和系统响应时间，引入粒子群算法对初始权值进行优化，最后应用Matlab软件对改进后的PID神经元网络算法进行仿真。仿真结果表明，该方法具有较好的控制性能。

Simulition of Multivariable Control Systems Based on Improved PID Neural Network

PID neural network has dynamic characteristics, and it can achieve better results than traditional PID control method in system control, but its learning algorithm is a gradient learning algorithm. The initial weights are random obtained. In order to improve the speed and system response time of the control system, we introduce the particle swarm algorithm to optimize the initial weights, and use Matlab software to simulate the improved PID neural network algorithm. The result of simulation reveals that the proposed method is better than traditional PID neural network control performance.